PDF to Excel Converter
Extract tables & data from any PDF into editable spreadsheets — including scanned image-based PDFs via built-in OCR. Zero uploads, 100% private.
PDF → Excel
Extract tables & data into editable spreadsheets
Drop your PDF here
Digital or scanned — tables & data extracted to Excel
🔒 Processed entirely in your browser — never uploaded
PDF to Excel: Extract Tables from Any PDF — Including Scanned Documents
Financial statements, research data, inventory reports, government statistics — enormous amounts of the world's most valuable data is locked inside PDF documents. This tool liberates that data using two distinct extraction pipelines: a spatial text engine for digital PDFs, and a full OCR engine for scanned image-based PDFs, giving you a clean Excel spreadsheet from any source document.
Why Is Data Trapped in PDFs?
PDF was designed as a presentation format, not a data format. A table in a PDF is not a table in any meaningful computational sense — it is a collection of text characters positioned at specific coordinates, visually arranged to look like rows and columns but with no inherent relationship between them. Scanned PDFs are even harder: they are photographs of pages, with no text data at all — every character must be recognized from pixels before any extraction can occur.
This is why copying and pasting from a PDF into Excel typically produces a mess. Recovering tabular structure requires analyzing the spatial positions of text on the page and inferring the row-column layout from those positions — or, for scanned PDFs, first recognizing the text with OCR and then applying the same spatial analysis.
Financial Statements
Balance sheets, income statements, and cash flow reports arrive as PDFs. Extracting to Excel enables ratio analysis, trend modeling, and period comparison.
Research & Survey Data
Academic papers, market research reports, and survey summaries include data tables that researchers need in spreadsheet form for statistical analysis.
Inventory & Logistics
Supplier price lists, inventory counts, and shipping manifests distributed as PDFs need to be in Excel for stock management and ERP system imports.
Government & Public Data
Census data, economic statistics, and regulatory filings released as PDFs — often scanned — contain tables analysts need to extract for visualization and analysis.
Two Extraction Pipelines — Digital and Scanned
This tool uses two completely different engines depending on whether OCR is enabled:
For text-based PDFs (created digitally). Mozilla's PDF.js returns every text item with precise X/Y coordinates. Items are grouped by Y-position into rows, then clustered by X-position into columns, reconstructing the original table structure spatially.
For image-based PDFs (scanned documents). Each page is rendered to a canvas at high resolution, then Tesseract.js performs optical character recognition to extract text with position data, which is then fed into the same spatial table reconstruction engine.
The Three Extraction Modes
Groups text items by Y-position into rows, then clusters X-positions to identify column boundaries. Each value is placed in the correct cell based on its spatial location. Best for formal financial reports and structured data tables.
Extracts all text from every page into a single-column text block, one page per sheet. Useful when you need all content faithfully preserved for further manual processing.
Treats each line as a row and splits values by the detected delimiter. Ideal for PDFs that contain data already formatted as delimited values exported from databases.
What to Expect: Accuracy and Limitations
- Simple tables with clear structure
- Financial statements with aligned numbers
- Digitally created PDFs
- Clearly printed scanned documents (with OCR)
- Single-column data lists
- CSV/delimited data saved as PDF
- Complex multi-level nested headers
- Tables spanning multiple pages
- Low-quality or skewed scans (OCR)
- Tables with merged cells
- Mixed table/text content documents
- Extremely blurry or degraded scans
- Handwritten data
- Password-protected PDFs
- Highly graphical PDF designs
- Non-Latin scripts (without OCR language config)
Tips for the Best Results
- Try without OCR first: Open your PDF and try to select text. If text highlights, it's digital — use the default mode (faster, more accurate). If nothing selects, enable OCR.
- Use Smart mode for formal tables: For financial reports and structured lists, Smart mode produces the best organized output with correct row-column structure.
- OCR works best on clean scans: The cleaner and higher resolution the scan, the better OCR accuracy. 300 DPI or higher produces excellent results.
- CSV mode for exported data: If your PDF was originally exported from a database containing comma/pipe-separated rows, CSV mode correctly splits each value into its own cell.
- Post-process in Excel: Use Excel's Text to Columns (Data → Text to Columns) to further split content. AutoFilter and Remove Duplicates are useful for cleaning extracted data.
- OCR is slower: Tesseract processes each page individually. A 10-page scanned PDF may take 30–90 seconds. Progress is shown throughout.
Frequently Asked Questions
Everything you need to know about converting PDF to Excel with OCR support.
Is this tool completely free?
Yes — completely free, no usage limits, no account required. OCR via Tesseract.js is also completely free and runs in your browser.
Is my PDF uploaded to a server?
No. Everything runs inside your browser — PDF.js, Tesseract OCR, and SheetJS all execute locally on your device's CPU. Your file never leaves your machine at any point.
When should I enable OCR mode?
Enable OCR when your PDF is a scanned document or contains image-based text. Test first: open your PDF and try to select text with your cursor. If nothing highlights, the PDF is image-based and needs OCR. If text selects, leave OCR off — the direct PDF.js extraction is faster and more accurate for digital PDFs.
Why does my spreadsheet look disorganized?
PDF does not store true table structure — it only stores text at positions. Extraction infers table structure from those positions. Try switching extraction modes: Smart Table Detection works best for formal tables, while CSV mode works better for line-by-line delimited data. For scanned PDFs, OCR accuracy also affects the result.
How long does OCR take?
OCR processes each page individually. A single page takes 3–8 seconds on a modern desktop. A 10-page scanned PDF typically takes 30–80 seconds. Progress is shown per page throughout. For large documents, a desktop or laptop will be significantly faster than mobile.
What is the difference between .xlsx and .csv?
XLSX supports multiple sheets, cell formatting (bold, colors, column widths), and opens natively in Excel, LibreOffice, and Google Sheets. CSV is a plain text format — one sheet, no formatting, universally compatible with all software. Choose .xlsx for rich structured output; choose .csv for database imports or system integrations.
Can I convert password-protected PDFs?
No. First remove the password: open the PDF with your password in Adobe Reader or your browser's PDF viewer, then use Print → Save as PDF to create an unprotected version. Convert the unprotected copy here.
What Excel versions open the output?
The .xlsx output is compatible with Microsoft Excel 2007 through Microsoft 365, LibreOffice Calc, Google Sheets (via Drive upload), Apple Numbers, and any software supporting .xlsx. The .csv output is universally compatible with all spreadsheet software.
Does the tool work on mobile phones?
Yes. The tool works on iOS Safari and Android Chrome. OCR is significantly slower on mobile due to processor speed — for large scanned PDFs, a desktop or laptop produces much faster results.
Which browsers are supported?
Chrome 70+, Firefox 65+, Safari 12+, Edge 79+, and Opera. All libraries require modern JavaScript APIs standard since 2018. Internet Explorer is not supported. Chrome or Firefox on desktop provides the best performance for OCR workloads.